Nov 26, 2013 - M2M Prime Business Criteria. Reliability ... Problems of ZigBee-like Solutions. Interference in ... GSM â
The Internet of Things (IoT) A Brave New World Jesus Alonso-Zarate, PhD Head of M2M Department Centre Tecnològic de Telecomunicacions de Catalunya (CTTC) Barcelona, Spain November 2013 © 2013 J. Alonso-Zarate
1
Disclaimer
Many third party copyrighted material is reused within this talk under the 'fair use' approach, for sake of educational purpose only, and very limited edition. As a consequence, the current slide set presentation usage is restricted, and is falling under usual copyrights usage. Thank you for your understanding!
© 2013 J. Alonso-Zarate
2
The Vision
© 2013 J. Alonso-Zarate
3
Internet of Things is all about… Things (Useful) People Connectivity
Interoperability Energy Efficiency Added Value Big Data Business Privacy and Security Well-Being © 2013 J. Alonso-Zarate
4
IoT can become a new revolution
http://www.gereports.com/new_industrial_internet_service_ technologies_from_ge_could_eliminate_150_billion_in_waste/ © 2013 J. Alonso-Zarate
5
Applications
© 2013 J. Alonso-Zarate
© http://postscapes.com/internet-of-things-examples/
6
Smart Cities
© 2013 J. Alonso-Zarate
7
Smart City Mashup Platforms Machine-to-Machine
Sensor Streams (Real Time)
BIG DATA Analytics
Improve Efficiency
Human-to-Machine
Crowdsourcing
W
Offer New Services
KNW Information-to-Machine
INFO
Internet (Open Data)
© 2013 J. Alonso-Zarate
Power Applications DATA
8
Technological Challenges
Device Domain
Network Domain M2M Communications
Applications Domain
© 2013 J. Alonso-Zarate
9
M2M Communications: The Access Network
© 2013 J. Alonso-Zarate
10
A General View (Wireless)
© 2013 J. Alonso-Zarate
11
M2M Prime Business Criteria Availability
Standardized Cellular Proprietary Cellular Low Power WLAN
Reliability
Bluetooth LE
Zigbee-like
Wired M2M
Availability = coverage, roaming, mobility, critical mass in rollout, etc. Reliability = resilience to interference, throughput guarantees, low outages, etc. (Total Cost of Ownership = CAPEX, OPEX.) © 2013 J. Alonso-Zarate
12
IEEE
3GPP
IETF
Standardized M2M Protocol Stack
© 2013 J. Alonso-Zarate
13
2.1
Capillary Networks (Zigbee-like and LP-WIFI)
© 2013 J. Alonso-Zarate
14
Problems of ZigBee-like Solutions Interference in ISM
No Global Infrastructure
2bn Wifi Devices
Lack of Interoperability
Higher Total Cost
WPA2/PSK/TLS/SSL © 2013 J. Alonso-Zarate
15
Advantages of Low-Power WiFi Ubiquitous Infrastructure
Vibrant Standard
300 members
2bn Wifi Devices Source: Wireless Broadband Access (WBA), Informa, Nov. 2011
Interference Management
Sound Security
NAV Medium Reservation
WPA2/PSK/TLS/SSL © 2013 J. Alonso-Zarate
16
LP-Wifi vs ZigBee Capillary M2M
10x
“Low-power Wi-Fi provides a significant improvement over typical Wi-Fi on both latency and energy consumption counts.” “LP-Wifi consumes approx the same as 6LoWPAN for small packets but is much better for large packets.”
[© IEEE, from “Feasibility of Wi-Fi Enabled Sensors for Internet of Things,” by Serbulent Tozlu (2011) © 2013 J. Alonso-Zarate
17
Low-Power Wifi Eco-System [examples]
© 2013 J. Alonso-Zarate
18
Low-Power Wifi Products [© Gainspan]
© 2013 J. Alonso-Zarate
19
Cellular M2M
© 2013 J. Alonso-Zarate
20
Advantages of Cellular M2M Ubiquitous Coverage
Mobility & Roaming
Interference Control
© 2013 J. Alonso-Zarate
21
However… ITU-R req. for IMT-Advanced
Means to achieve higher data rates: More spectrum, more efficient RRM, smaller cells
2G
2.5G
3G
3.5G
4GExabyte = 10^185G
Source: NEC – Andreas Maeder, Feb 2012 © 2013 J. Alonso-Zarate
22
Key Technical Novelties
Cellular Networks have been designed for humans!
Accommodation of M2M requires paradigm shift:
There will be a lot of M2M nodes More and more applications are delay-intolerant, mainly control There will be little traffic per node, and mainly in the uplink Nodes need to run autonomously for a long time Automated security & trust mechanisms
… and all this without jeopardizing current cellular services!
© 2013 J. Alonso-Zarate
23
Challenges for Mobile Operators
Lack of M2M experience
mobile operators are experts in human-to-human (H2H) M2M is a new market and a mental shift is required
High operational costs
the network has to be dimensioned for a number of devices that just transmit few bits of information from time to time
Low Average Revenue Per User (ARPU) Fragmentation and complexity of applications Lack of standardization (so far) Competition from other (emerging) technologies
© 2013 J. Alonso-Zarate
Low Power Wide Area (LPWA ) Technologies
24
2.2.1
M2M in Current Cellular Networks How suitable are current technologies for M2M?
© 2013 J. Alonso-Zarate
25
GSM – PHY and MAC Layers
PHY Layer Carrier Frequency: 900 MHz, 1.8 GHz, and others. Simple Power Management: 8 power classes; min 20 mW = 13 dBm (2dB power control steps) Modulation with Constant envelope (good for PA) PHY Data Rates: 9.6 Kbps Low Complexity MAC Layer Duplexing: FDD FDMA/TDMA + ALOHA-based access Traffic Type: Voice, Data, 160 7-bit SMS.
© 2013 J. Alonso-Zarate
26
Beyond GSM – GPRS & EDGE
GPRS = GSM + …
© 2013 J. Alonso-Zarate
… more time slots for users + … adaptive coding schemes
EDGE = GPRS + …
… 8PSK modulation scheme
27
UMTS – PHY and MAC Layers
PHY Layer Carrier Frequency: 2Ghz, and others. Instantaneous Power Management CDMA Modulation: Variable envelope PHY Data Rates: >100 kbit/s packet switched Medium Complexity
MAC Layer Duplexing: FDD FDMA/CDMA (256 codes) + ALOHA-based access
Traffic type: conversational, streaming, interactive, background.
© 2013 J. Alonso-Zarate
28
LTE & LTE-A
LTE (Release 8 and 9)
OFDMA (downlink) + SC-FDMA (uplink) Robust to multipath Flexible spectrum allocation (adjusting number of subcarriers) Efficient receiver implementations Simple MIMO implementation in frequency domain freq. diversity gain Quicker RTT & throughput Both TDD and FDD duplexing modes Variable bandwidth (1.4 to 20MHz) Spectral Efficiency (x3) Simplified Architecture lower CAPEX and OPEX More User Capacity (x10)
LTE-A (Releases 10-11-12): LTE + new features + M2M support
© 2013 J. Alonso-Zarate
DL: 1Gbps, UL: 500 Mbps.
29
Source: presentation by Thierry Lestable, Sagemcom, BEFEMTO Winter School, Barcelona, Feb. 2012.
Key Limitations of LTE & LTE-A FDD: 2x70MHz TDD: 50MHz
FDD: 2x35MHz China Mobile Genius Brand CSL Ltd …
AWS TeliaSonera Vodafone O2 …
Hong-Kong
Major TD-LTE Market (incl. India)
Verizon AT&T metroPCS NTT DoCoMo
Refarming and extensions are still to come… Digital Dividend Fragmentation & Harmonization of Spectrum is a critical problem! © 2013 J. Alonso-Zarate
30
Key Limitations of LTE & LTE-A
Not efficient for small data transmission Scheduled Radio access
Random access and more flexibility
Device cost issues
Scalable bandwidth Data rate (overdesigned UE categories) Transmit power (max. 23dBm) Half Duplex operation (simpler device) RF chains with 2 antennas Signal processing accuracy
Overload issues big number of devices High mobility support
Source: IP-FP7-258512 EXALTED D3.1 © 2013 J. Alonso-Zarate
31
2.2.2
Standardization Activities
http://www.3gpp.org/ftp/Information/WORK_PLAN/Description_Releases/M2M_20130924.zip
© 2013 J. Alonso-Zarate
32
Standards for Cellular M2M
Industry has become more active in standardizing M2M:
Because of the market demand Essential for long term development of technology For interoperability of networks Ability to “roam” M2M services over international frontiers
Due to potentially heavy use of M2M devices and thus high loads onto networks, interest from:
© 2013 J. Alonso-Zarate
ETSI TC M2M and recently oneM2M Partnership Project 3GPP (GSM, EDGE GPRS, UMTS, HSPA, LTE) IEEE 802.16 (WiMAX)
33
ETSI: Inverting the pipes
© ETSI © 2013 J. Alonso-Zarate
34
ETSI: Functional Architecture Service Capabilities shared by different applications
Core Network: - IP Connectivity - Interconnection with other networks - Roaming with other core networks - Service and Network control
Registration, Authentication, Authorization, Management, Provisioning
© ETSI © 2013 J. Alonso-Zarate
35
Creation of oneM2M Partnership project
China
USA
Korea
© 2013 J. Alonso-Zarate
Japan
36
ETSI Smart Card Platform (2000)
SIM: Subscriber Identity Module more than 4B in circulation
Evolution to UICC (Universal Integrated Circuit Card) CPU, RAM, ROM, EEPROM, I/O.
June 2012, 4th Form Factor (nano-SIM , from Apple)
Technical Specs TS 102 221 v11.0.0 (2012-06)
http://www.etsi.org/deliver/etsi_ts/102200_102299/102221/11.00.00_60/ts_102221v110000p.pdf
Data of the SIM: ICCID: identifier IMSI Authentication key Location Area Identity User data: Contacts and SMS Embedded SIMs for M2M
© 2013 J. Alonso-Zarate
37
3GPP: M2M Features
A feature is a system optimization possibility Different requirements different optimizations Offered on a per subscription basis: Low Mobility Time Controlled Time Tolerant Small Data Transmissions Mobile originated only Infrequent Mobile Terminated MTC Monitoring
© 2013 J. Alonso-Zarate
Priority Alarm Message (PAM) Secure Connection Location Specific Trigger Infrequent transmission Group Based features Policing Addressing
38
2.3
Specific M2M Architectures & Performance
© 2013 J. Alonso-Zarate
39
2.3.1
Tools to Play Around with the IoT
© 2013 J. Alonso-Zarate
40
Tools
Math Computer-based Simulators (MATLAB, Ns-3 Simulator) Cellular (LENA simulator) Capillary Networks (Zigbee-Like, WiFi, etc.)
Testbeds Arduino, WizziMotes, PanStamps, Raspberry Pi, OpenMAC, WARP, Digi Xbee Modules, etc. Not that much in LTE and cellular
Free M2M Platforms in the Cloud
© 2013 J. Alonso-Zarate
Xybely, Thingspeak, etc.
41
2.3.2
A Possible M2M Architecture EXALTED was an FP7 funded IP Project (#258512)
© 2013 J. Alonso-Zarate
42
At A Glance: EXALTED
Expanding LTE for Devices
ICT EXALTED Expanding LTE for Devices
Project Coordinator Djelal Raouf Sagemcom SAS Tel: +33 (0)1 57 61 20 08 Fax: +33 (0)1 57 61 39 09 Email:
[email protected] Project website: www.ict-exalted.eu Partners: Vodafone Group Services Limited (UK), Vodafone Group Services GmbH (DE), Gemalto (FR), Ericsson d.o.o. Serbia (RS), AlcatelLucent (DE), Telekom Srbija (RS), Commissariat à l’énergie atomique et aux energies alternatives (FR), TST Sistemas S.A. (ES), University of Surrey (UK), Centre Tecnològic de Telecomunicacions de Catalunya (ES), TUD Vodafone Chair (DE), University of Piraeus Research Center (GR), Vidavo SA (GR) Duration: Sept. 2010 – Feb. 2013 Funding scheme: IP Total Cost: €11m EC Contribution: €7.4m Contract Number: INFSO-ICT258512
© 2013 J. Alonso-Zarate
43
Some Specific EXALTED Solutions
LTE-M: Backwards compatible solution
PHY Layer
Use of TDD
Generalized FDM for Uplink (beyond SC-FDMA)
Use of LDPC Codes instead of Turbo Codes
Low-Complexity MIMO
Cooperative Relaying
© 2013 J. Alonso-Zarate
44
Some Specific EXALTED Solutions
Higher Layers:
Optimal scheduling with energy-harvesting
Scheduling for heterogeneous traffic (event-driven or periodic)
Discontinuous Reception (DRX): Duty Cycle
Use of RACH for small data transmission, adding CDMA for data collision recovery.
© 2013 J. Alonso-Zarate
45
2.3.3
High Number of Devices: The access to the network The Random Access Channel (RACH) of LTE
© 2013 J. Alonso-Zarate
46
Examples of RACH Configuration 1.08 MHz (6 PRBs)
Less resources for data!
© 2013 J. Alonso-Zarate
47
The Procedure 64 Orthogonal Preambles Not all available for random access LIMITATIONS Preamble transmission based on FS-ALOHA HARQ
Preamble Collisions Collisions in Message 3 Lack of resources for Msg3
http://prezi.com/likwdk7dksmj/ns-3-lte-random-access-procedure/ © 2013 J. Alonso-Zarate
48
Blocking Probability: bulk arrival
MAX retx = 10
© 2013 J. Alonso-Zarate
49
Energy Consumption: bulk arrival
MAX retx = 10
© 2013 J. Alonso-Zarate
50
Blocking Probability: more retx
© 2013 J. Alonso-Zarate
51
Energy Consumption: more retx
© 2013 J. Alonso-Zarate
52
Delay: more retx
© 2013 J. Alonso-Zarate
53
RACH Improvements
A. Laya, L. Alonso, J. Alonso-Zarate “Is the Random Access Channel of LTE and LTE-A Suitable for M2M Communications? A Survey of Alternatives” IEEE Tutorials and Survey Communications Magazine Special Issue on Machine-to-Machine Communications. Publication Date: January 2014.
© 2013 J. Alonso-Zarate
54
Distributed Queuing as a Solution
Go away from ALOHA-type access
Contention-Tree Algorithm (CTA)
Distributed Queuing Random Access Protocol (DQRAP)
Performance independent of number of users
Simple operation with two logical queues
One for transmission of successful access requests
One to solve contention
Downlink Control Channel of LTE ideal for implementation J. Alonso-Zárate, E. Kartsakli, A. Cateura, C. Verikoukis, and L. Alonso. “A Near-Optimum Cross-Layered Distributed Queuing Protocol for Wireless LAN,” IEEE Wireless Communication Magazine. Special Issue on MAC protocols for WLAN, vol. 15, no. 1, pp. 48-55, February 2008.
© 2013 J. Alonso-Zarate
55
2.3.4
Coexistence of Machines and Humans
© 2013 J. Alonso-Zarate
56
Impact of MTC onto HTC
Dropping probabilities, duty cycle and delay for 30min access case: 0.7
HTC MTC Method 1 Method 2
0.6
Drop probability
0.5
0.4
0.3
0.2
0.1
0.0 0
20000
40000
60000
80000
100000
MTC number © Kan Zheng, Jesus Alonso-Zarate, and Mischa Dohler
© 2013 J. Alonso-Zarate
57
2.3.5
Limitations of the RLC layer
© 2013 J. Alonso-Zarate
58
RRC State Transitions
© 2013 J. Alonso-Zarate
59
Impact of MTC onto HTC
“A Close Examination of Performance and Power Characteristics of 4G LTE Networks,” by J. Huang, F. Qian, A. Gerber. MobiSYS’12.
screen
LTE
3G WiFi
© 2013 J. Alonso-Zarate
60
2.3.6
Optimizing LTE for Small Data Transmissions
K. Wang, J. Alonso-Zarate, M. Dohler, “Energy-Efficiency of LTE for Small Data Machine-to-Machine Communications,” in proc. of the IEEE ICC 2013, Budapest, Hungary, June 2013.
© 2013 J. Alonso-Zarate
61
Transport Block Size vs. MCS
15
Transmit Power (dBm)
10
5
0
-5
-10
0
5
10
15
20
25
MCS Index
© 2013 J. Alonso-Zarate
62
Example
© 2013 J. Alonso-Zarate
63
The Optimal MCS Transport Block Utilization Rate (Packet size/ TBS)
1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 10 bytes packet 35 bytes packet 60 bytes packet
0.1 0
0
5
10
15
20
25
MCS Index
© 2013 J. Alonso-Zarate
64
The Optimal MCS 9
6
x 10
10 bytes packet 35 bytes packet 60 bytes packet
Energy Efficiency η (bits/Joule)
5
4
3
2
1
0
0
5
10
15
20
25
MCS Index
© 2013 J. Alonso-Zarate
65
The Market
© 2013 J. Alonso-Zarate
66
Number of Connections
These numbers include in-building network infrastructure such as residential modems/ routers .
Global machine-to-machine connections 2011-22 Source: Machina Research 2012 © 2013 J. Alonso-Zarate
67
M2M Technologies
Global machine-to-machine connections 2011-22 Source: Machina Research 2012 © 2013 J. Alonso-Zarate
68
A USD 1.2 trillion opportunity by 2022
Total M2M revenue will grow from USD200 billion in 2011 to USD1.2 trillion in 2022, a CAGR of 18% Total revenue includes:
Device costs where connectivity is integral to the device Module costs where devices can optionally have connectivity enabled Monthly subscription, connectivity and traffic fees
Global machine-to-machine connections 2011-22 Source: Machina Research 2012 © 2013 J. Alonso-Zarate
69
Defining New Business Models
Global machine-to-machine connections 2011-22 Source: Machina Research 2012 © 2013 J. Alonso-Zarate
70
Humans in the Loop
© 2013 J. Alonso-Zarate
71
Machines can do what we cannot
© 2013 J. Alonso-Zarate
72
Machines can do better than us…
© 2013 J. Alonso-Zarate
73
An Example:26 November 2013 Employees against use of automatic supermarket check-out
© 2013 J. Alonso-Zarate
74
Productivity Paradox
PRODUCTIVITY
EMPLOYMENT AVERAGE WORKER (income)
Erik Brynjolfsson (director of the MIT Center for Digital Busines) says: “there is no economic law that says that technological progress needs to benefit everybody, or even that it needs to benefit a majority of people. It’s entirely possible for technology to advance, to make the pie bigger, and yet for some people to get a smaller share of that pie.” http://www.mckinsey.com/insights/high_tech_telecoms_internet/charting_technologys_new_directions_a_conversation_with_mits_erik_brynjolfsson
© 2013 J. Alonso-Zarate
75
Decoupling Technology & Economy
VS
© 2013 J. Alonso-Zarate
76
Conclusions
© 2013 J. Alonso-Zarate
77
Conclusions The
IoT is around the corner. There is a huge market opportunity Many challenges ahead:
Technology is (almost) here, but must be optimized. • • • •
Zero-Power operation Dense deployments Long distances Simplicity
New Business Models are needed. Politicians must push for it (smart cities) Humans in the loop
© 2013 J. Alonso-Zarate
78
IoT World Forum, October 2013
The IoT is nascent, and its value needs to be defined 90% of the business is comprised of start-ups
Industrial internet + consumers-facing opportunities. Consumers will be a source of innovation in the IoT.
No single company will build the IoT Major companies will need to find ways to engage with and enable these builders.
Arduino, Raspberry Pi, Thingsquare, Libelium’s Cooking Hacks , Smart Citizen Kit, TheThings.io, Electric Imp to Telefonica’s recently announced Thinking Things, and Intel’s Galileo, among others.
The IoT will not rest on one killer app, but on openness and interoperability
Source: http://www.claropartners.com/the-emerging-iot-business-landscape-insights-from-the-iot-world-forum-2/ © 2013 J. Alonso-Zarate
79
Final Take-Away Message Henry Ford “If I had asked people what they wanted, they would have said… A FASTER HORSE!”
© 2013 J. Alonso-Zarate
80
THANKS Jesus Alonso-Zarate,
[email protected] Twitter: @jalonsozarate www.jesusalonsozarate.com
© 2013 J. Alonso-Zarate
81